Communication devices such as those in a monitoring system for a building facility or a plant facility are connected as nodes via a communication network. The communication system such as a monitoring system is typically deployed in an Ethernet network via which nodes communicate with each other according to a set of protocols.
Ring topology has been proposed for connecting each node in the monitoring system for industrial facilities. Compared to a star topology in which each node is connected via separate wiring, a ring topology particularly a single ring configuration alleviates the workload involved with the wiring work and maintenance. A ring network may provide redundant communication paths to avoid communication cut-off in the event of a failure.
FIG. 1 is an example of a configuration of a typical ring network 100 in which nodes are provided with a network control function that performs spanning tree control methods such as Spanning Tree Protocol (STP) and Rapid Spanning Tree Protocol (RSTP). In FIG. 1, the path from node N1 which acts as a master node to node N2 could be either counterclockwise or clockwise. If the cost of the counterclockwise direction is lower than that of the clockwise direction, then the counterclockwise path will be selected as an active communication path 120. Accordingly, the path from the node N2 to node N3 becomes an unnecessary path. One port of nodes on this unnecessary path is blocked to establish a redundant communication path 140. Consequently, a loop configuration is logically prevented by setting a port on the redundant communication path 140 in a blocking state.
In the event that a link between two nodes fails or a node of the ring network malfunctions, the blocked port of the node on that redundant communication path 140 is unblocked in order to establish a new communication path. In the example shown in FIG. 1, if a failure occurs at point P1, node N2 may unblock point P2 to form a new path from node N1 to node N2 through node N3.
However, spanning tree control methods that involve path recalculation may consume considerable computing resources. This may lead to a decrease in system performance and an increase in restoration time of data communication. Also, the amount of time consumed for path recalculation is proportional to the number of the nodes in the ring network so the total number of the nodes may significantly affect the amount of time for switching from a failed path to a new path. Moreover, with spanning tree control methods, the ring network recovered from failure could not activate data communication until each node learns a new communication path notified by the master node.
It is therefore an object of the present invention to address one or more of the above issues.